ci: fix

chore: package.json repository url
chore: Remove readers package from llamaindex (#1649 )
2026-07-15 14:55:41 -04:00 · 2025-02-12 21:12:41 +08:00 · 2025-02-12 21:05:34 +08:00 · 2025-02-12 17:16:41 +07:00 · 2025-02-12 15:10:11 +07:00 · 2025-02-12 12:45:26 +07:00
174 changed files with 5190 additions and 7570 deletions
@@ -0,0 +1,16 @@
+---
+"@llamaindex/milvus": minor
+"@llamaindex/qdrant": minor
+"@llamaindex/next-node-runtime-test": minor
+"@llamaindex/azure": minor
+"@llamaindex/cloudflare-hono": minor
+"@llamaindex/anthropic": minor
+"@llamaindex/llamaindex-test": minor
+"llamaindex": minor
+"@llamaindex/core": minor
+"@llamaindex/doc": minor
+"@llamaindex/examples": minor
+"@llamaindex/e2e": minor
+---
+
+Remove re-exports from llamaindex main package
@@ -0,0 +1,5 @@
+---
+"@llamaindex/doc": patch
+---
+
+docs: update chat engine docs
@@ -0,0 +1,5 @@
+---
+"@llamaindex/doc": patch
+---
+
+docs: update workflow doc
@@ -0,0 +1,6 @@
+---
+"@llamaindex/core": patch
+"llamaindex": patch
+---
+
+feat: asChatEngine function for index
@@ -0,0 +1,7 @@
+---
+"llamaindex": minor
+"@llamaindex/cloudflare-hono": patch
+"@llamaindex/examples": patch
+---
+
+Remove deprecated ServiceContext
@@ -0,0 +1,8 @@
+---
+"llamaindex": minor
+"@llamaindex/doc": minor
+"@llamaindex/examples": minor
+"@llamaindex/unit-test": minor
+---
+
+Remove readers package from llamaindex
@@ -25,4 +25,4 @@ jobs:
        run: pnpm run build

      - name: Pre Release
-        run: pnpx pkg-pr-new publish ./packages/* ./packages/providers/*
+        run: pnpx pkg-pr-new publish --pnpm ./packages/* ./packages/providers/* ./packages/providers/storage/*
@@ -83,11 +83,6 @@ jobs:
        run: pnpm install
      - name: Build
        run: pnpm run build
-      - name: Use Build For Examples
-        run: |
-          pnpm link ../packages/llamaindex/
-          cd readers && pnpm link ../../packages/llamaindex/
-        working-directory: ./examples
      - name: Run Type Check
        run: pnpm run type-check
      - name: Run Circular Dependency Check
@@ -1,3 +1 @@
-pnpm format
-pnpm lint
-npx lint-staged
+pnpm run lint-staged
@@ -0,0 +1 @@
+LlamaIndexTS
@@ -14,5 +14,6 @@
  "[json]": {
    "editor.defaultFormatter": "esbenp.prettier-vscode"
  },
-  "prettier.prettierPath": "./node_modules/prettier"
+  "prettier.prettierPath": "./node_modules/prettier",
+  "prettier.configPath": "prettier.config.mjs"
 }
@@ -65,44 +65,18 @@ yarn add llamaindex

 See our official document: <https://ts.llamaindex.ai/docs/llamaindex/getting_started/>

-### Tips when using in non-Node.js environments
+### Adding provider packages

-When you are importing `llamaindex` in a non-Node.js environment(such as Vercel Edge, Cloudflare Workers, etc.)
-Some classes are not exported from top-level entry file.
+In most cases, you'll also need to install provider packages to use LlamaIndexTS. These are for adding AI models, file readers for ingestion or storing documents, e.g. in vector databases.

-The reason is that some classes are only compatible with Node.js runtime,(e.g. `PDFReader`) which uses Node.js specific APIs(like `fs`, `child_process`, `crypto`).
+For example, to use the OpenAI LLM, you would install the following package:

-If you need any of those classes, you have to import them instead directly though their file path in the package.
-Here's an example for importing the `PineconeVectorStore` class:
-
-```typescript
-import { PineconeVectorStore } from "llamaindex/vector-store/PineconeVectorStore";
+```shell
+npm install @llamaindex/openai
+pnpm install @llamaindex/openai
+yarn add @llamaindex/openai
 ```

-As the `PDFReader` is not working with the Edge runtime, here's how to use the `SimpleDirectoryReader` with the `LlamaParseReader` to load PDFs:
-
-```typescript
-import { SimpleDirectoryReader } from "llamaindex/readers/SimpleDirectoryReader";
-import { LlamaParseReader } from "llamaindex/readers/LlamaParseReader";
-
-export const DATA_DIR = "./data";
-
-export async function getDocuments() {
-  const reader = new SimpleDirectoryReader();
-  // Load PDFs using LlamaParseReader
-  return await reader.loadData({
-    directoryPath: DATA_DIR,
-    fileExtToReader: {
-      pdf: new LlamaParseReader({ resultType: "markdown" }),
-    },
-  });
-}
-```
-
-> _Note_: Reader classes have to be added explictly to the `fileExtToReader` map in the Edge version of the `SimpleDirectoryReader`.
-
-You'll find a complete example with LlamaIndexTS here: https://github.com/run-llama/create_llama_projects/tree/main/nextjs-edge-llamaparse
-
 ## Playground

 Check out our NextJS playground at https://llama-playground.vercel.app/. The source is available at https://github.com/run-llama/ts-playground
@@ -57,4 +57,3 @@ In this example, the Context-Aware Agent uses the retriever to fetch relevant co
 ## Available Context-Aware Agents

 - `OpenAIContextAwareAgent`: A context-aware agent using OpenAI's models.
- `AnthropicContextAwareAgent`: A context-aware agent using Anthropic's models.
@@ -15,7 +15,7 @@ In LlamaIndex, an agent is a semi-autonomous piece of software powered by an LLM
 You'll need to have a recent version of [Node.js](https://nodejs.org/en) installed. Then you can install LlamaIndex.TS by running

 ```bash
-npm install llamaindex
+npm install llamaindex @llamaindex/openai @llamaindex/readers @llamaindex/huggingface
 ```

 ## Choose your model
@@ -40,7 +40,7 @@ We'll be bringing in `SimpleDirectoryReader`, `HuggingFaceEmbedding`, `VectorSto
 import { FunctionTool, QueryEngineTool, Settings, VectorStoreIndex } from "llamaindex";
 import { OpenAI, OpenAIAgent } from "@llamaindex/openai";
 import { HuggingFaceEmbedding } from "@llamaindex/huggingface";
-import { SimpleDirectoryReader } from "llamaindex";
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
 ```

 ### Add an embedding model
@@ -10,7 +10,7 @@ import { Accordion, Accordions } from 'fumadocs-ui/components/accordion';
 <Accordions>
 	<Accordion title="Install @llamaindex/readers">

-		If you want to only use reader modules, you can install `@llamaindex/readers`
+		If you want to use the reader module, you need to install `@llamaindex/readers`

 		<Tabs groupId="install-llamaindex" items={["npm", "yarn", "pnpm"]} persist>
 			```shell tab="npm"
@@ -31,72 +31,73 @@ import { Accordion, Accordions } from 'fumadocs-ui/components/accordion';

 We offer readers for different file formats.

-<Tabs groupId="llamaindex-or-readers" items={["llamaindex", "@llamaindex/readers"]} persist>
-	```ts twoslash tab="llamaindex"
-	import { CSVReader } from '@llamaindex/readers/csv'
-	import { PDFReader } from '@llamaindex/readers/pdf'
-	import { JSONReader } from '@llamaindex/readers/json'
-	import { MarkdownReader } from '@llamaindex/readers/markdown'
-	import { HTMLReader } from '@llamaindex/readers/html'
-	// you can find more readers in the documentation
-	```
-
-	```ts twoslash tab="@llamaindex/readers"
-	import { CSVReader } from '@llamaindex/readers/csv'
-	import { PDFReader } from '@llamaindex/readers/pdf'
-	import { JSONReader } from '@llamaindex/readers/json'
-	import { MarkdownReader } from '@llamaindex/readers/markdown'
-	import { HTMLReader } from '@llamaindex/readers/html'
-	// you can find more readers in the documentation
-	```
-
-</Tabs>
+```ts twoslash 
+import { CSVReader } from '@llamaindex/readers/csv'
+import { PDFReader } from '@llamaindex/readers/pdf'
+import { JSONReader } from '@llamaindex/readers/json'
+import { MarkdownReader } from '@llamaindex/readers/markdown'
+import { HTMLReader } from '@llamaindex/readers/html'
+// you can find more readers in the documentation
+```

 ## SimpleDirectoryReader

 `SimpleDirectoryReader` is the simplest way to load data from local files into LlamaIndex.

-<Tabs groupId="llamaindex-or-readers" items={["llamaindex", "@llamaindex/readers"]} persist>
+```ts twoslash 
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";

-	```ts twoslash tab="llamaindex"
-	import { SimpleDirectoryReader } from "llamaindex";
-
-	const reader = new SimpleDirectoryReader()
-	const documents = await reader.loadData("./data")
-	//    ^?
+const reader = new SimpleDirectoryReader()
+const documents = await reader.loadData("./data")
+//    ^?


-	const texts = documents.map(doc => doc.getText())
-	//    ^?
-	```
-
-	```ts twoslash tab="@llamaindex/readers"
-	import { SimpleDirectoryReader } from "llamaindex";
-
-	const reader = new SimpleDirectoryReader()
-	const documents = await reader.loadData("./data")
-	//    ^?
+const texts = documents.map(doc => doc.getText())
+//    ^?
+```


-	const texts = documents.map(doc => doc.getText())
-	//    ^?
-	```
+## Tips when using in non-Node.js environments
+
+When using `@llamaindex/readers` in a non-Node.js environment (such as Vercel Edge, Cloudflare Workers, etc.)
+Some classes are not exported from top-level entry file.
+
+The reason is that some classes are only compatible with Node.js runtime, (e.g. `PDFReader`) which uses Node.js specific APIs (like `fs`, `child_process`, `crypto`).
+
+If you need any of those classes, you have to import them instead directly through their file path in the package.
+
+As the `PDFReader` is not working with the Edge runtime, here's how to use the `SimpleDirectoryReader` with the `LlamaParseReader` to load PDFs:
+
+```typescript
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
+import { LlamaParseReader } from "@llamaindex/cloud";
+
+export const DATA_DIR = "./data";
+
+export async function getDocuments() {
+  const reader = new SimpleDirectoryReader();
+  // Load PDFs using LlamaParseReader
+  return await reader.loadData({
+    directoryPath: DATA_DIR,
+    fileExtToReader: {
+      pdf: new LlamaParseReader({ resultType: "markdown" }),
+    },
+  });
+}
+```
+
+> _Note_: Reader classes have to be added explicitly to the `fileExtToReader` map in the Edge version of the `SimpleDirectoryReader`.
+
+You'll find a complete example with LlamaIndexTS here: https://github.com/run-llama/create_llama_projects/tree/main/nextjs-edge-llamaparse

-</Tabs>

 ## Load file natively using Node.js Customization Hooks

 We have a helper utility to allow you to import a file in Node.js script.

-<Tabs groupId="llamaindex-or-readers" items={["llamaindex", "@llamaindex/readers"]} persist>
-	```shell tab="llamaindex"
-	node --import llamaindex/register ./script.js
-	```
-
-	```shell tab="@llamaindex/readers"
-	node --import @llamaindex/readers/node ./script.js
-	```
-</Tabs>
+```shell 
+node --import @llamaindex/readers/node ./script.js
+```

 ```ts
 import csv from './path/to/data.csv';
@@ -12,9 +12,26 @@ const chatEngine = new ContextChatEngine({ retriever });
 const response = await chatEngine.chat({ message: query });
 ```

+In short, you can use the chat engine by calling `index.asChatEngine()`. It will return a `ContextChatEngine` to start chatting.
+
+```typescript
+const chatEngine = index.asChatEngine();
+```
+
+You can also pass in options to the chat engine.
+
+```typescript
+const chatEngine = index.asChatEngine({
+  similarityTopK: 5,
+  systemPrompt: "You are a helpful assistant.",
+});
+```
+
+
 The `chat` function also supports streaming, just add `stream: true` as an option:

 ```typescript
+const chatEngine = index.asChatEngine();
 const stream = await chatEngine.chat({ message: query, stream: true });
 for await (const chunk of stream) {
  process.stdout.write(chunk.response);
@@ -34,7 +34,7 @@ import {
  Settings,
 } from "llamaindex";
 import { OpenAI } from "@llamaindex/openai";
-import { SimpleDirectoryReader } from "llamaindex";
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
 ```

 ## Loading Data
@@ -124,7 +124,7 @@ import {
  Settings,
 } from "llamaindex";
 import { OpenAI } from "@llamaindex/openai";
-import { SimpleDirectoryReader } from "llamaindex";
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";

 Settings.llm = new OpenAI();
 Settings.nodeParser = new SentenceSplitter({
@@ -13,6 +13,22 @@ When a step function is added to a workflow, you need to specify the input and o

 You can create a `Workflow` to do anything! Build an agent, a RAG flow, an extraction flow, or anything else you want.

+import { Tab, Tabs } from "fumadocs-ui/components/tabs";
+
+<Tabs groupId="install" items={["npm", "yarn", "pnpm"]} persist>
+	```shell tab="npm"
+	npm install @llamaindex/workflow
+	```
+
+	```shell tab="yarn"
+	yarn add @llamaindex/workflow
+	```
+
+	```shell tab="pnpm"
+	pnpm add @llamaindex/workflow
+	```
+</Tabs>
+
 ## Getting Started

 As an illustrative example, let's consider a naive workflow where a joke is generated and then critiqued.
@@ -34,51 +50,59 @@ Events are user-defined classes that extend `WorkflowEvent` and contain arbitrar
 ```typescript
 const llm = new OpenAI();
 ...
-const jokeFlow = new Workflow({ verbose: true });
+const jokeFlow = new Workflow<unknown, string, string>();
 ```

-Our workflow is implemented by initiating the `Workflow` class. For simplicity, we created a `OpenAI` llm instance.
+Our workflow is implemented by initiating the `Workflow` class with three generic types: the context type (unknown), input type (string), and output type (string). The context type is `unknown`, as we're not using a shared context in this example.
+
+For simplicity, we created an `OpenAI` llm instance that we're using for inference in our workflow.

 ### Workflow Entry Points

 ```typescript
-const generateJoke = async (_context: Context, ev: StartEvent) => {
-  const prompt = `Write your best joke about ${ev.data.input}.`;
+const generateJoke = async (_: unknown, ev: StartEvent<string>) => {
+  const prompt = `Write your best joke about ${ev.data}.`;
  const response = await llm.complete({ prompt });
  return new JokeEvent({ joke: response.text });
 };
 ```

-Here, we come to the entry-point of our workflow. While events are user-defined, there are two special-case events, the `StartEvent` and the `StopEvent`. Here, the `StartEvent` signifies where to send the initial workflow input.
+Here, we come to the entry-point of our workflow. While events are user-defined, there are two special-case events, the `StartEvent` and the `StopEvent`. These events are predefined, but we can specify the payload type using generic types. We're using `StartEvent<string>` to indicate that we're going to send an input of type string.

-The `StartEvent` is a bit of a special object since it can hold arbitrary attributes. Here, we accessed the topic with `ev.data.input`.
-
-At this point, you may have noticed that we haven't explicitly told the workflow what events are handled by which steps.
-
-To do so, we use the `addStep` method which adds a step to the workflow. The first argument is the event type that the step will handle, and the second argument is the previously defined step function:
+To add this step to the workflow, we use the `addStep` method with an object specifying the input and output event types:

 ```typescript
-jokeFlow.addStep(StartEvent, generateJoke);
+jokeFlow.addStep(
+  {
+    inputs: [StartEvent<string>],
+    outputs: [JokeEvent],
+  },
+  generateJoke
+);
 ```

 ### Workflow Exit Points

 ```typescript
-const critiqueJoke = async (_context: Context, ev: JokeEvent) => {
+const critiqueJoke = async (_: unknown, ev: JokeEvent) => {
  const prompt = `Give a thorough critique of the following joke: ${ev.data.joke}`;
  const response = await llm.complete({ prompt });
-  return new StopEvent({ result: response.text });
+  return new StopEvent(response.text);
 };
 ```

-Here, we have our second, and last step, in the workflow. We know its the last step because the special `StopEvent` is returned. When the workflow encounters a returned `StopEvent`, it immediately stops the workflow and returns whatever the result was.
+Here, we have our second and last step in the workflow. We know it's the last step because the special `StopEvent` is returned. When the workflow encounters a returned `StopEvent`, it immediately stops the workflow and returns the result. Note that we're using the generic type `StopEvent<string>` to indicate that we're returning a string.

-In this case, the result is a string, but it could be a map, array, or any other object.
-
-Don't forget to add the step to the workflow:
+Add this step to the workflow:

 ```typescript
-jokeFlow.addStep(JokeEvent, critiqueJoke);
+jokeFlow.addStep(
+  {
+    inputs: [JokeEvent],
+    outputs: [StopEvent<string>],
+  },
+  critiqueJoke
+);
 ```

 ### Running the Workflow
@@ -90,42 +114,25 @@ console.log(result.data.result);

 Lastly, we run the workflow. The `.run()` method is async, so we use await here to wait for the result.

-### Validating Workflows
+## Working with Shared Context/State

-To tell the workflow what events are produced by each step, you can optionally provide a third argument to `addStep` to specify the output event type:
+Optionally, you can choose to use a shared context between steps by specifying a context type when creating the workflow. Here's an example where multiple steps access a shared state:

 ```typescript
-jokeFlow.addStep(StartEvent, generateJoke, { outputs: JokeEvent });
-jokeFlow.addStep(JokeEvent, critiqueJoke, { outputs: StopEvent });
-```
+import { HandlerContext } from "@llamaindex/workflow";

-To validate a workflow, you need to call the `validate` method:
+type MyContextData = {
+  query: string;
+  intermediateResults: any[];
+}

-```typescript
-jokeFlow.validate();
-```
-
-To automatically validate a workflow when you run it, you can set the `validate` flag to `true` at initialization:
-
-```typescript
-const jokeFlow = new Workflow({ verbose: true, validate: true });
-```
-
-## Working with Global Context/State
-
-Optionally, you can choose to use global context between steps. For example, maybe multiple steps access the original `query` input from the user. You can store this in global context so that every step has access.
-
-```typescript
-import { Context } from "llamaindex";
-
-const query = async (context: Context, ev: MyEvent) => {
+const query = async (context: HandlerContext<MyContextData>, ev: MyEvent) => {
  // get the query from the context
-  const query = context.get("query");
+  const query = context.data.query;
  // do something with context and event
  const val = ...
-  const result = ...
  // store in context
-  context.set("key", val);
+  context.data.intermediateResults.push(val);

  return new StopEvent({ result });
 };
@@ -138,28 +145,15 @@ The context does more than just hold data, it also provides utilities to buffer
 For example, you might have a step that waits for a query and retrieved nodes before synthesizing a response:

 ```typescript
-const synthesize = async (context: Context, ev: QueryEvent | RetrieveEvent) => {
-  const events = context.collectEvents(ev, [QueryEvent | RetrieveEvent]);
-  if (!events) {
-    return;
-  }
-  const prompt = events
-    .map((event) => {
-      if (event instanceof QueryEvent) {
-        return `Answer this query using the context provided: ${event.data.query}`;
-      } else if (event instanceof RetrieveEvent) {
-        return `Context: ${event.data.context}`;
-      }
-      return "";
-    })
-    .join("\n");
-
+const synthesize = async (context: Context, ev1: QueryEvent, ev2: RetrieveEvent) => {
+  const subPrompts = [`Answer this query using the context provided: ${ev1.data.query}`, `Context: ${ev2.data.context}`];
+  const prompt = subPrompts.join("\n");
  const response = await llm.complete({ prompt });
  return new StopEvent({ result: response.text });
 };
 ```

-Using `ctx.collectEvents()` we can buffer and wait for ALL expected events to arrive. This function will only return events (in the requested order) once all events have arrived.
+Passing multiple events, we can buffer and wait for ALL expected events to arrive. The receiving step function will only be called once all events have arrived.

 ## Manually Triggering Events

@@ -1 +1,2 @@
 logs
+.temp
@@ -17,23 +17,21 @@ app.post("/llm", async (c) => {

  const { message } = await c.req.json();

-  const { extractText } = await import("@llamaindex/core/utils");
-
  const {
+    extractText,
    QueryEngineTool,
-    serviceContextFromDefaults,
    VectorStoreIndex,
-    OpenAIAgent,
    Settings,
-    OpenAI,
-    OpenAIEmbedding,
+    SentenceSplitter,
  } = await import("llamaindex");

-  const { PineconeVectorStore } = await import(
-    "llamaindex/vector-store/PineconeVectorStore"
+  const { OpenAIAgent, OpenAI, OpenAIEmbedding } = await import(
+    "@llamaindex/openai"
  );

-  const llm = new OpenAI({
+  const { PineconeVectorStore } = await import("@llamaindex/pinecone");
+
+  Settings.llm = new OpenAI({
    model: "gpt-4o-mini",
    apiKey: c.env.OPENAI_API_KEY,
  });
@@ -43,8 +41,7 @@ app.post("/llm", async (c) => {
    apiKey: c.env.OPENAI_API_KEY,
  });

-  const serviceContext = serviceContextFromDefaults({
-    llm,
+  Settings.nodeParser = new SentenceSplitter({
    chunkSize: 8191,
    chunkOverlap: 0,
  });
@@ -53,7 +50,7 @@ app.post("/llm", async (c) => {
    namespace: "8xolsn4ulEQGdhnhP76yCzfLHdOZ",
  });

-  const index = await VectorStoreIndex.fromVectorStore(store, serviceContext);
+  const index = await VectorStoreIndex.fromVectorStore(store);

  const retriever = index.asRetriever({
    similarityTopK: 3,
@@ -9,6 +9,8 @@
  },
  "dependencies": {
    "llamaindex": "workspace:*",
+    "@llamaindex/huggingface": "workspace:*",
+    "@llamaindex/readers": "workspace:*",
    "next": "15.0.3",
    "react": "18.3.1",
    "react-dom": "18.3.1"
@@ -1,13 +1,13 @@
 "use server";
+import { HuggingFaceEmbedding } from "@llamaindex/huggingface";
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
 import {
  OpenAI,
  OpenAIAgent,
  QueryEngineTool,
  Settings,
-  SimpleDirectoryReader,
  VectorStoreIndex,
 } from "llamaindex";
-import { HuggingFaceEmbedding } from "llamaindex/embeddings/HuggingFaceEmbedding";

 Settings.llm = new OpenAI({
  apiKey: process.env.NEXT_PUBLIC_OPENAI_KEY ?? "FAKE_KEY_TO_PASS_TESTS",
@@ -9,6 +9,7 @@
    "start": "waku start"
  },
  "dependencies": {
+    "@llamaindex/env": "workspace:*",
    "llamaindex": "workspace:*",
    "react": "19.0.0-rc-5c56b873-20241107",
    "react-dom": "19.0.0-rc-5c56b873-20241107",
@@ -1,13 +1,14 @@
 "use server";
+import { fs } from "@llamaindex/env";
 import { BaseQueryEngine, Document, VectorStoreIndex } from "llamaindex";
-import { readFile } from "node:fs/promises";
+
 let _queryEngine: BaseQueryEngine;

 async function lazyLoadQueryEngine() {
  if (!_queryEngine) {
    const path = "node_modules/llamaindex/examples/abramov.txt";

-    const essay = await readFile(path, "utf-8");
+    const essay = await fs.readFile(path, "utf-8");

    // Create Document object with essay
    const document = new Document({ text: essay, id_: path });
@@ -1,7 +1,7 @@
+import { Anthropic, AnthropicAgent } from "@llamaindex/anthropic";
 import { extractText } from "@llamaindex/core/utils";
 import { consola } from "consola";
-import { Anthropic, FunctionTool, Settings, type LLM } from "llamaindex";
-import { AnthropicAgent } from "llamaindex/agent/anthropic";
+import { FunctionTool, Settings, type LLM } from "llamaindex";
 import { ok } from "node:assert";
 import { beforeEach, test } from "node:test";
 import { getWeatherTool, sumNumbersTool } from "./fixtures/tools.js";
@@ -1,6 +1,7 @@
+import { ClipEmbedding } from "@llamaindex/clip";
 import type { LoadTransformerEvent } from "@llamaindex/env/multi-model";
 import { setTransformers } from "@llamaindex/env/multi-model";
-import { ClipEmbedding, ImageNode, Settings } from "llamaindex";
+import { ImageNode, Settings } from "llamaindex";
 import assert from "node:assert";
 import { type Mock, test } from "node:test";

@@ -0,0 +1,86 @@
+import { execSync } from "node:child_process";
+import { mkdir, rm, writeFile } from "node:fs/promises";
+import { resolve } from "node:path";
+import { test } from "node:test";
+import { testRootDir } from "./utils.js";
+
+await test("cjs/esm dual module check", async (t) => {
+  const esmImports = `import fs from 'node:fs/promises'
+import { Document, MetadataMode, VectorStoreIndex } from 'llamaindex'
+import { OpenAIEmbedding } from '@llamaindex/openai'
+import { Settings } from '@llamaindex/core/global'`;
+  const cjsRequire = `const fs = require('fs').promises
+const { Document, MetadataMode, VectorStoreIndex } = require('llamaindex')
+const { OpenAIEmbedding } = require('@llamaindex/openai')
+const { Settings } = require('@llamaindex/core/global')`;
+  const mainCode = `
+async function main() {
+  Settings.embedModel = new OpenAIEmbedding({
+    model: 'text-embedding-3-small',
+    apiKey: '${process.env.OPENAI_API_KEY}',
+  })
+  const model = Settings.embedModel
+  if (model == null) {
+    process.exit(-1)
+  }
+}
+main().catch(console.error)`;
+  t.before(async () => {
+    await mkdir(resolve(testRootDir, ".temp"), {
+      recursive: true,
+      mode: 0o755,
+    });
+  });
+
+  t.after(async () => {
+    await rm(resolve(testRootDir, ".temp"), {
+      recursive: true,
+      force: true,
+    });
+  });
+
+  await t.test("cjs", async () => {
+    const cjsCode = `${cjsRequire}\n${mainCode}`;
+    const filePath = resolve(
+      testRootDir,
+      ".temp",
+      `${crypto.randomUUID()}.cjs`,
+    );
+    await writeFile(filePath, cjsCode, "utf-8");
+
+    execSync(`${process.argv[0]} ${filePath}`, {
+      cwd: process.cwd(),
+    });
+  });
+
+  await t.test("esm", async () => {
+    const esmCode = `${esmImports}\n${mainCode}`;
+    const filePath = resolve(
+      testRootDir,
+      ".temp",
+      `${crypto.randomUUID()}.mjs`,
+    );
+    await writeFile(filePath, esmCode, "utf-8");
+
+    execSync(`${process.argv[0]} ${filePath}`, {
+      cwd: process.cwd(),
+    });
+  });
+
+  const specialConditions = ["edge-light", "workerd", "react-server"];
+  for (const condition of specialConditions) {
+    await t.test(condition, async () => {
+      const esmCode = `${esmImports}\n${mainCode}`;
+      const filePath = resolve(
+        testRootDir,
+        ".temp",
+        `${crypto.randomUUID()}.mjs`,
+      );
+      await writeFile(filePath, esmCode, "utf-8");
+
+      execSync(`${process.argv[0]} ${filePath} -C ${condition}`, {
+        cwd: process.cwd(),
+      });
+    });
+  }
+});
@@ -1,6 +1,6 @@
+import { PGVectorStore } from "@llamaindex/postgres";
 import { config } from "dotenv";
 import { Document, VectorStoreQueryMode } from "llamaindex";
-import { PGVectorStore } from "llamaindex/vector-store/PGVectorStore";
 import assert from "node:assert";
 import { test } from "node:test";
 import pg from "pg";
@@ -1,10 +1,8 @@
 import { Document, MetadataMode } from "@llamaindex/core/schema";
+import { OpenAIEmbedding } from "@llamaindex/openai";
+import { PineconeVectorStore } from "@llamaindex/pinecone";
 import { config } from "dotenv";
-import {
-  OpenAIEmbedding,
-  PineconeVectorStore,
-  VectorStoreIndex,
-} from "llamaindex";
+import { VectorStoreIndex } from "llamaindex";
 import assert from "node:assert";
 import { test } from "node:test";

@@ -14,6 +14,10 @@
    "@llamaindex/env": "workspace:*",
    "@llamaindex/ollama": "workspace:*",
    "@llamaindex/openai": "workspace:*",
+    "@llamaindex/pinecone": "workspace:*",
+    "@llamaindex/postgres": "workspace:*",
+    "@llamaindex/clip": "workspace:*",
+    "@llamaindex/anthropic": "workspace:*",
    "@types/node": "^22.9.0",
    "@types/pg": "^8.11.8",
    "@huggingface/transformers": "^3.0.2",
@@ -1,9 +1,6 @@
 import { OpenAIAgent } from "@llamaindex/openai";
-import {
-  QueryEngineTool,
-  SimpleDirectoryReader,
-  VectorStoreIndex,
-} from "llamaindex";
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
+import { QueryEngineTool, VectorStoreIndex } from "llamaindex";

 async function main() {
  // Load the documents
@@ -1,9 +1,9 @@
 import { OpenAIAgent } from "@llamaindex/openai";
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
 import {
  FunctionTool,
  MetadataMode,
  NodeWithScore,
-  SimpleDirectoryReader,
  VectorStoreIndex,
 } from "llamaindex";

@@ -1,9 +1,6 @@
 import { OpenAIAgent } from "@llamaindex/openai";
-import {
-  QueryEngineTool,
-  SimpleDirectoryReader,
-  VectorStoreIndex,
-} from "llamaindex";
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
+import { QueryEngineTool, VectorStoreIndex } from "llamaindex";

 async function main() {
  // Load the documents
@@ -1,5 +1,5 @@
 import { AstraDBVectorStore } from "@llamaindex/astra";
-import { VectorStoreIndex, serviceContextFromDefaults } from "llamaindex";
+import { VectorStoreIndex } from "llamaindex";

 const collectionName = "movie_reviews";

@@ -8,8 +8,7 @@ async function main() {
    const astraVS = new AstraDBVectorStore({ contentKey: "reviewtext" });
    await astraVS.connect(collectionName);

-    const ctx = serviceContextFromDefaults();
-    const index = await VectorStoreIndex.fromVectorStore(astraVS, ctx);
+    const index = await VectorStoreIndex.fromVectorStore(astraVS);

    const retriever = await index.asRetriever({ similarityTopK: 20 });

@@ -0,0 +1,15 @@
+import { Document, KeywordTableIndex } from "llamaindex";
+import essay from "../essay";
+
+async function main() {
+  const document = new Document({ text: essay });
+  const index = await KeywordTableIndex.fromDocuments([document]);
+  const chatEngine = index.asChatEngine();
+
+  const response = await chatEngine.chat({
+    message: "What is Harsh Mistress?",
+  });
+  console.log(response.message.content);
+}
+
+main().catch(console.error);
@@ -0,0 +1,17 @@
+import { Document, SummaryIndex, SummaryRetrieverMode } from "llamaindex";
+import essay from "../essay";
+
+async function main() {
+  const document = new Document({ text: essay });
+  const index = await SummaryIndex.fromDocuments([document]);
+  const chatEngine = index.asChatEngine({
+    mode: SummaryRetrieverMode.LLM,
+  });
+
+  const response = await chatEngine.chat({
+    message: "Summary about the author",
+  });
+  console.log(response.message.content);
+}
+
+main().catch(console.error);
@@ -0,0 +1,15 @@
+import { Document, VectorStoreIndex } from "llamaindex";
+import essay from "../essay";
+
+async function main() {
+  const document = new Document({ text: essay });
+  const index = await VectorStoreIndex.fromDocuments([document]);
+  const chatEngine = index.asChatEngine({ similarityTopK: 5 });
+
+  const response = await chatEngine.chat({
+    message: "What did I work on in February 2021?",
+  });
+  console.log(response.message.content);
+}
+
+main().catch(console.error);
@@ -1,4 +1,4 @@
-import { SimpleDirectoryReader } from "llamaindex";
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";

 function callback(
  category: string,
@@ -30,13 +30,12 @@ async function main() {

  // Split text and create embeddings. Store them in a VectorStoreIndex
  // var storageContext = await storageContextFromDefaults({});
-  // var serviceContext = serviceContextFromDefaults({});
  // const docStore = storageContext.docStore;

  // for (const doc of documents) {
  //   docStore.setDocumentHash(doc.id_, doc.hash);
  // }
-  // const nodes = serviceContext.nodeParser.getNodesFromDocuments(documents);
+  // const nodes = Settings.nodeParser.getNodesFromDocuments(documents);
  // console.log(nodes);

  //
@@ -1,9 +1,9 @@
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
 import {
  ImageDocument,
  JinaAIEmbedding,
  similarity,
  SimilarityType,
-  SimpleDirectoryReader,
 } from "llamaindex";
 import path from "path";

@@ -1,4 +1,5 @@
-import { Settings, SimpleDirectoryReader, VectorStoreIndex } from "llamaindex";
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
+import { Settings, VectorStoreIndex } from "llamaindex";
 import path from "path";
 import { getStorageContext } from "./storage";

@@ -1,17 +1,16 @@
-import { storageContextFromDefaults } from "llamaindex";
+import { ClipEmbedding } from "@llamaindex/clip";
+import { path } from "@llamaindex/env";
+import { SimpleVectorStore, storageContextFromDefaults } from "llamaindex";

 // set up store context with two vector stores, one for text, the other for images
 export async function getStorageContext() {
  return await storageContextFromDefaults({
    persistDir: "storage",
-    storeImages: true,
-    // if storeImages is true, the following vector store will be added
-    // vectorStores: {
-    //   IMAGE: SimpleVectorStore.fromPersistDir(
-    //     `${persistDir}/images`,
-    //     fs,
-    //     new ClipEmbedding(),
-    //   ),
-    // },
+    vectorStores: {
+      IMAGE: await SimpleVectorStore.fromPersistDir(
+        path.join("storage", "images"),
+        new ClipEmbedding(),
+      ),
+    },
  });
 }
@@ -1,5 +1,4 @@
-import { OllamaEmbedding } from "@llamaindex/ollama";
-import { Ollama } from "llamaindex/llm/ollama";
+import { Ollama, OllamaEmbedding } from "@llamaindex/ollama";

 (async () => {
  const llm = new Ollama({
@@ -1,63 +1,64 @@
 {
  "name": "@llamaindex/examples",
-  "private": true,
  "version": "0.1.3",
+  "private": true,
+  "scripts": {
+    "lint": "eslint .",
+    "start": "tsx ./starter.ts"
+  },
  "dependencies": {
    "@ai-sdk/openai": "^1.0.5",
    "@azure/cosmos": "^4.1.1",
    "@azure/identity": "^4.4.1",
    "@azure/search-documents": "^12.1.0",
-    "@llamaindex/vercel": "^0.0.10",
-    "@llamaindex/workflow": "^0.0.10",
+    "@llamaindex/anthropic": "workspace:* || ^0.0.33",
+    "@llamaindex/astra": "workspace:* || ^0.0.4",
+    "@llamaindex/azure": "workspace:* || ^0.0.4",
+    "@llamaindex/chroma": "workspace:* || ^0.0.4",
+    "@llamaindex/clip": "workspace:* || ^0.0.35",
+    "@llamaindex/cloud": "workspace:* || ^2.0.24",
+    "@llamaindex/cohere": "workspace:* || ^0.0.4",
+    "@llamaindex/deepinfra": "workspace:* || ^0.0.35",
+    "@llamaindex/env": "workspace:* || ^0.1.27",
+    "@llamaindex/google": "workspace:* || ^0.0.6",
+    "@llamaindex/groq": "workspace:* || ^0.0.50",
+    "@llamaindex/huggingface": "workspace:* || ^0.0.35",
+    "@llamaindex/milvus": "workspace:* || ^0.0.4",
+    "@llamaindex/mistral": "workspace:* || ^0.0.4",
+    "@llamaindex/mixedbread": "workspace:* || ^0.0.4",
+    "@llamaindex/mongodb": "workspace:* || ^0.0.4",
+    "@llamaindex/node-parser": "workspace:* || ^0.0.24",
+    "@llamaindex/ollama": "workspace:* || ^0.0.39",
+    "@llamaindex/openai": "workspace:* || ^0.1.51",
+    "@llamaindex/pinecone": "workspace:* || ^0.0.4",
+    "@llamaindex/portkey-ai": "workspace:* || ^0.0.32",
+    "@llamaindex/postgres": "workspace:* || ^0.0.32",
+    "@llamaindex/qdrant": "workspace:* || ^0.0.4",
+    "@llamaindex/readers": "workspace:* || ^1.0.25",
+    "@llamaindex/replicate": "workspace:* || ^0.0.32",
+    "@llamaindex/upstash": "workspace:* || ^0.0.4",
+    "@llamaindex/vercel": "workspace:* || ^0.0.10",
+    "@llamaindex/vllm": "workspace:* || ^0.0.21",
+    "@llamaindex/weaviate": "workspace:* || ^0.0.4",
+    "@llamaindex/workflow": "workspace:* || ^0.0.10",
    "@notionhq/client": "^2.2.15",
    "@pinecone-database/pinecone": "^4.0.0",
    "@vercel/postgres": "^0.10.0",
    "ai": "^4.0.0",
+    "ajv": "^8.17.1",
    "commander": "^12.1.0",
    "dotenv": "^16.4.5",
    "js-tiktoken": "^1.0.14",
-    "llamaindex": "^0.8.37",
+    "llamaindex": "workspace:* || ^0.8.37",
    "mongodb": "6.7.0",
    "postgres": "^3.4.4",
-    "ajv": "^8.17.1",
-    "wikipedia": "^2.1.2",
-    "@llamaindex/openai": "workspace:*",
-    "@llamaindex/cloud": "workspace:*",
-    "@llamaindex/anthropic": "workspace:*",
-    "@llamaindex/clip": "workspace:*",
-    "@llamaindex/azure": "workspace:*",
-    "@llamaindex/deepinfra": "workspace:*",
-    "@llamaindex/groq": "workspace:*",
-    "@llamaindex/huggingface": "workspace:*",
-    "@llamaindex/node-parser": "workspace:*",
-    "@llamaindex/ollama": "workspace:*",
-    "@llamaindex/portkey-ai": "workspace:*",
-    "@llamaindex/readers": "workspace:*",
-    "@llamaindex/replicate": "workspace:*",
-    "@llamaindex/vllm": "workspace:*",
-    "@llamaindex/postgres": "workspace:*",
-    "@llamaindex/astra": "workspace:*",
-    "@llamaindex/milvus": "workspace:*",
-    "@llamaindex/chroma": "workspace:*",
-    "@llamaindex/mongodb": "workspace:*",
-    "@llamaindex/pinecone": "workspace:*",
-    "@llamaindex/qdrant": "workspace:*",
-    "@llamaindex/upstash": "workspace:*",
-    "@llamaindex/weaviate": "workspace:*",
-    "@llamaindex/google": "workspace:*",
-    "@llamaindex/mistral": "workspace:*",
-    "@llamaindex/mixedbread": "workspace:*",
-    "@llamaindex/cohere": "workspace:*"
+    "wikipedia": "^2.1.2"
  },
  "devDependencies": {
    "@types/node": "^22.9.0",
    "tsx": "^4.19.0",
    "typescript": "^5.7.2"
  },
-  "scripts": {
-    "lint": "eslint .",
-    "start": "tsx ./starter.ts"
-  },
  "stackblitz": {
    "startCommand": "npm start"
  }
@@ -1,11 +1,8 @@
 // load-docs.ts
 import { PineconeVectorStore } from "@llamaindex/pinecone";
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
 import fs from "fs/promises";
-import {
-  SimpleDirectoryReader,
-  storageContextFromDefaults,
-  VectorStoreIndex,
-} from "llamaindex";
+import { storageContextFromDefaults, VectorStoreIndex } from "llamaindex";

 async function getSourceFilenames(sourceDir: string) {
  return await fs
@@ -19,9 +19,9 @@
    "start:obsidian": "node --import tsx ./src/obsidian.ts"
  },
  "dependencies": {
-    "@llamaindex/readers": "*",
-    "llamaindex": "*",
-    "@llamaindex/cloud": "*"
+    "@llamaindex/cloud": "workspace:* || ^2.0.24",
+    "@llamaindex/readers": "workspace:* || ^1.0.25",
+    "llamaindex": "workspace:* || ^0.8.37"
  },
  "devDependencies": {
    "@types/node": "^22.9.0",
@@ -1,10 +1,10 @@
-import { TextFileReader } from "@llamaindex/readers/text";
-import type { Document, Metadata } from "llamaindex";
 import {
  FILE_EXT_TO_READER,
-  FileReader,
  SimpleDirectoryReader,
-} from "llamaindex";
+} from "@llamaindex/readers/directory";
+import { TextFileReader } from "@llamaindex/readers/text";
+import type { Document, Metadata } from "llamaindex";
+import { FileReader } from "llamaindex";

 class ZipReader extends FileReader {
  loadDataAsContent(fileContent: Uint8Array): Promise<Document<Metadata>[]> {
@@ -1,5 +1,6 @@
 import { LlamaParseReader } from "@llamaindex/cloud";
-import { SimpleDirectoryReader, VectorStoreIndex } from "llamaindex";
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
+import { VectorStoreIndex } from "llamaindex";

 async function main() {
  const reader = new SimpleDirectoryReader();
@@ -1,6 +1,4 @@
-import { SimpleDirectoryReader } from "llamaindex";
-// or
-// import { SimpleDirectoryReader } from 'llamaindex'
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";

 const reader = new SimpleDirectoryReader();
 const documents = await reader.loadData("../data");
@@ -1,9 +1,9 @@
 import { OpenAI } from "@llamaindex/openai";
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
 import {
  RouterQueryEngine,
  SentenceSplitter,
  Settings,
-  SimpleDirectoryReader,
  SummaryIndex,
  VectorStoreIndex,
 } from "llamaindex";
@@ -14,6 +14,7 @@ import {
  MetadataIndexFieldType,
 } from "@llamaindex/azure";
 import { OpenAI, OpenAIEmbedding } from "@llamaindex/openai";
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
 import dotenv from "dotenv";
 import {
  Document,
@@ -22,7 +23,6 @@ import {
  Metadata,
  NodeWithScore,
  Settings,
-  SimpleDirectoryReader,
  storageContextFromDefaults,
  TextNode,
  VectorStoreIndex,
@@ -1,10 +1,7 @@
 // load-docs.ts
 import { PGVectorStore } from "@llamaindex/postgres";
-import {
-  SimpleDirectoryReader,
-  storageContextFromDefaults,
-  VectorStoreIndex,
-} from "llamaindex";
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
+import { storageContextFromDefaults, VectorStoreIndex } from "llamaindex";
 import fs from "node:fs/promises";

 async function getSourceFilenames(sourceDir: string) {
@@ -1,6 +1,6 @@
+import { PGVectorStore } from "@llamaindex/postgres";
 import dotenv from "dotenv";
 import { Document, VectorStoreQueryMode } from "llamaindex";
-import { PGVectorStore } from "llamaindex/vector-store/PGVectorStore";
 import postgres from "postgres";

 dotenv.config();
@@ -1,5 +1,5 @@
+import { PGVectorStore } from "@llamaindex/postgres";
 import { VectorStoreIndex } from "llamaindex";
-import { PGVectorStore } from "llamaindex/vector-store/PGVectorStore";

 async function main() {
  // eslint-disable-next-line @typescript-eslint/no-require-imports
@@ -1,10 +1,7 @@
 import { PGVectorStore } from "@llamaindex/postgres";
+import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
 import dotenv from "dotenv";
-import {
-  SimpleDirectoryReader,
-  storageContextFromDefaults,
-  VectorStoreIndex,
-} from "llamaindex";
+import { storageContextFromDefaults, VectorStoreIndex } from "llamaindex";

 dotenv.config();

@@ -1,8 +1,8 @@
 // https://vercel.com/docs/storage/vercel-postgres/sdk
+import { PGVectorStore } from "@llamaindex/postgres";
 import { sql } from "@vercel/postgres";
 import dotenv from "dotenv";
 import { Document, VectorStoreQueryMode } from "llamaindex";
-import { PGVectorStore } from "llamaindex/vector-store/PGVectorStore";

 dotenv.config();

@@ -2,8 +2,9 @@
  "name": "@llamaindex/monorepo",
  "private": true,
  "scripts": {
-    "build": "turbo run build --filter=\"./packages/*\" --filter=\"./packages/providers/*\"",
-    "dev": "turbo run dev --filter=\"./packages/*\" --filter=\"./packages/providers/*\"",
+    "clean": "find . -type d \\( -name .turbo -o -name node_modules -o -name dist -o -name .next -o -name lib \\) -exec rm -rf {} +",
+    "build": "turbo run build --filter=\"./packages/*\" --filter=\"./packages/providers/**\"",
+    "dev": "turbo run dev --filter=\"./packages/*\" --filter=\"./packages/providers/**\"",
    "format": "prettier --ignore-unknown --cache --check .",
    "format:write": "prettier --ignore-unknown --write .",
    "lint": "turbo run lint",
@@ -15,7 +16,8 @@
    "release": "pnpm run build && changeset publish",
    "release-snapshot": "pnpm run build && changeset publish --tag snapshot",
    "new-version": "changeset version && pnpm format:write && pnpm run build",
-    "new-snapshot": "pnpm run build && changeset version --snapshot"
+    "new-snapshot": "pnpm run build && changeset version --snapshot",
+    "lint-staged": "lint-staged"
  },
  "devDependencies": {
    "@changesets/cli": "^2.27.5",
@@ -36,6 +38,10 @@
  },
  "packageManager": "pnpm@9.12.3",
  "lint-staged": {
-    "(!apps/docs/i18n/**/docusaurus-plugin-content-docs/current/api/*).{js,jsx,ts,tsx,md}": "prettier --write"
+    "*.{js,jsx,ts,tsx}": [
+      "prettier --check",
+      "eslint"
+    ],
+    "*.{json,md}": "prettier --check"
  }
 }
@@ -82,11 +82,8 @@
  }
  .background-gradient {
    background-color: #fff;
-    background-image: radial-gradient(
-        at 21% 11%,
-        rgba(186, 186, 233, 0.53) 0,
-        transparent 50%
-      ),
+    background-image:
+      radial-gradient(at 21% 11%, rgba(186, 186, 233, 0.53) 0, transparent 50%),
      radial-gradient(at 85% 0, hsla(46, 57%, 78%, 0.52) 0, transparent 50%),
      radial-gradient(at 91% 36%, rgba(194, 213, 255, 0.68) 0, transparent 50%),
      radial-gradient(at 8% 40%, rgba(251, 218, 239, 0.46) 0, transparent 50%);
@@ -1,6 +1,11 @@
 {
  "name": "@llamaindex/autotool",
  "type": "module",
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/run-llama/LlamaIndexTS.git",
+    "directory": "packages/autotool"
+  },
  "version": "5.0.37",
  "description": "auto transpile your JS function to LLM Agent compatible",
  "files": [
@@ -60,7 +60,7 @@
  },
  "repository": {
    "type": "git",
-    "url": "https://github.com/run-llama/LlamaIndexTS.git",
+    "url": "git+https://github.com/run-llama/LlamaIndexTS.git",
    "directory": "packages/cloud"
  },
  "devDependencies": {
@@ -34,7 +34,7 @@
  ],
  "repository": {
    "type": "git",
-    "url": "https://github.com/run-llama/LlamaIndexTS.git",
+    "url": "git+https://github.com/run-llama/LlamaIndexTS.git",
    "directory": "packages/community"
  },
  "scripts": {
@@ -386,7 +386,7 @@
  "repository": {
    "type": "git",
    "directory": "packages/core",
-    "url": "https://github.com/run-llama/LlamaIndexTS.git"
+    "url": "git+https://github.com/run-llama/LlamaIndexTS.git"
  },
  "devDependencies": {
    "@edge-runtime/vm": "^4.0.4",
@@ -20,6 +20,16 @@ import type {
 import { DefaultContextGenerator } from "./default-context-generator";
 import type { ContextGenerator } from "./type";

+export type ContextChatEngineOptions = {
+  retriever: BaseRetriever;
+  chatModel?: LLM | undefined;
+  chatHistory?: ChatMessage[] | undefined;
+  contextSystemPrompt?: ContextSystemPrompt | undefined;
+  nodePostprocessors?: BaseNodePostprocessor[] | undefined;
+  systemPrompt?: string | undefined;
+  contextRole?: MessageType | undefined;
+};
+
 /**
 * ContextChatEngine uses the Index to get the appropriate context for each query.
 * The context is stored in the system prompt, and the chat history is chunk,
@@ -35,15 +45,7 @@ export class ContextChatEngine extends PromptMixin implements BaseChatEngine {
    return this.memory.getMessages();
  }

-  constructor(init: {
-    retriever: BaseRetriever;
-    chatModel?: LLM | undefined;
-    chatHistory?: ChatMessage[] | undefined;
-    contextSystemPrompt?: ContextSystemPrompt | undefined;
-    nodePostprocessors?: BaseNodePostprocessor[] | undefined;
-    systemPrompt?: string | undefined;
-    contextRole?: MessageType | undefined;
-  }) {
+  constructor(init: ContextChatEngineOptions) {
    super();
    this.chatModel = init.chatModel ?? Settings.llm;
    this.memory = new ChatMemoryBuffer({ chatHistory: init?.chatHistory });
@@ -4,6 +4,9 @@ export {
  type NonStreamingChatEngineParams,
  type StreamingChatEngineParams,
 } from "./base";
-export { ContextChatEngine } from "./context-chat-engine";
+export {
+  ContextChatEngine,
+  type ContextChatEngineOptions,
+} from "./context-chat-engine";
 export { DefaultContextGenerator } from "./default-context-generator";
 export { SimpleChatEngine } from "./simple-chat-engine";
@@ -16,7 +16,6 @@ export const DEFAULT_DOC_STORE_PERSIST_FILENAME = "doc_store.json";
 export const DEFAULT_VECTOR_STORE_PERSIST_FILENAME = "vector_store.json";
 export const DEFAULT_GRAPH_STORE_PERSIST_FILENAME = "graph_store.json";
 export const DEFAULT_NAMESPACE = "docstore";
-export const DEFAULT_IMAGE_VECTOR_NAMESPACE = "images";
 //#endregion
 //#region llama cloud
 export const DEFAULT_PROJECT_NAME = "Default";
@@ -113,7 +113,7 @@
  ],
  "repository": {
    "type": "git",
-    "url": "https://github.com/run-llama/LlamaIndexTS.git",
+    "url": "git+https://github.com/run-llama/LlamaIndexTS.git",
    "directory": "packages/env"
  },
  "scripts": {
@@ -42,7 +42,7 @@
  ],
  "repository": {
    "type": "git",
-    "url": "https://github.com/run-llama/LlamaIndexTS.git",
+    "url": "git+https://github.com/run-llama/LlamaIndexTS.git",
    "directory": "packages/experimental"
  },
  "scripts": {
@@ -20,35 +20,11 @@
    "llamaindex"
  ],
  "dependencies": {
-    "@llamaindex/anthropic": "workspace:*",
-    "@llamaindex/clip": "workspace:*",
    "@llamaindex/cloud": "workspace:*",
    "@llamaindex/core": "workspace:*",
-    "@llamaindex/deepinfra": "workspace:*",
    "@llamaindex/env": "workspace:*",
-    "@llamaindex/groq": "workspace:*",
-    "@llamaindex/huggingface": "workspace:*",
    "@llamaindex/node-parser": "workspace:*",
-    "@llamaindex/ollama": "workspace:*",
    "@llamaindex/openai": "workspace:*",
-    "@llamaindex/portkey-ai": "workspace:*",
-    "@llamaindex/readers": "workspace:*",
-    "@llamaindex/replicate": "workspace:*",
-    "@llamaindex/vllm": "workspace:*",
-    "@llamaindex/postgres": "workspace:*",
-    "@llamaindex/azure": "workspace:*",
-    "@llamaindex/astra": "workspace:*",
-    "@llamaindex/milvus": "workspace:*",
-    "@llamaindex/chroma": "workspace:*",
-    "@llamaindex/mongodb": "workspace:*",
-    "@llamaindex/pinecone": "workspace:*",
-    "@llamaindex/qdrant": "workspace:*",
-    "@llamaindex/upstash": "workspace:*",
-    "@llamaindex/weaviate": "workspace:*",
-    "@llamaindex/google": "workspace:*",
-    "@llamaindex/mistral": "workspace:*",
-    "@llamaindex/mixedbread": "workspace:*",
-    "@llamaindex/cohere": "workspace:*",
    "@types/lodash": "^4.17.7",
    "@types/node": "^22.9.0",
    "ajv": "^8.17.1",
@@ -144,7 +120,7 @@
  ],
  "repository": {
    "type": "git",
-    "url": "https://github.com/run-llama/LlamaIndexTS.git",
+    "url": "git+https://github.com/run-llama/LlamaIndexTS.git",
    "directory": "packages/llamaindex"
  },
  "scripts": {
@@ -1,67 +0,0 @@
-import type { BaseEmbedding } from "@llamaindex/core/embeddings";
-import { PromptHelper } from "@llamaindex/core/indices";
-import type { LLM } from "@llamaindex/core/llms";
-import {
-  type NodeParser,
-  SentenceSplitter,
-} from "@llamaindex/core/node-parser";
-import { OpenAI, OpenAIEmbedding } from "@llamaindex/openai";
-
-/**
- * The ServiceContext is a collection of components that are used in different parts of the application.
- *
- * @deprecated This will no longer supported, please use `Settings` instead.
- */
-export interface ServiceContext {
-  llm: LLM;
-  promptHelper: PromptHelper;
-  embedModel: BaseEmbedding;
-  nodeParser: NodeParser;
-  // llamaLogger: any;
-}
-
-export interface ServiceContextOptions {
-  llm?: LLM;
-  promptHelper?: PromptHelper;
-  embedModel?: BaseEmbedding;
-  nodeParser?: NodeParser;
-  // NodeParser arguments
-  chunkSize?: number;
-  chunkOverlap?: number;
-}
-
-export function serviceContextFromDefaults(options?: ServiceContextOptions) {
-  const serviceContext: ServiceContext = {
-    llm: options?.llm ?? new OpenAI(),
-    embedModel: options?.embedModel ?? new OpenAIEmbedding(),
-    nodeParser:
-      options?.nodeParser ??
-      new SentenceSplitter({
-        chunkSize: options?.chunkSize,
-        chunkOverlap: options?.chunkOverlap,
-      }),
-    promptHelper: options?.promptHelper ?? new PromptHelper(),
-  };
-
-  return serviceContext;
-}
-
-export function serviceContextFromServiceContext(
-  serviceContext: ServiceContext,
-  options: ServiceContextOptions,
-) {
-  const newServiceContext = { ...serviceContext };
-  if (options.llm) {
-    newServiceContext.llm = options.llm;
-  }
-  if (options.promptHelper) {
-    newServiceContext.promptHelper = options.promptHelper;
-  }
-  if (options.embedModel) {
-    newServiceContext.embedModel = options.embedModel;
-  }
-  if (options.nodeParser) {
-    newServiceContext.nodeParser = options.nodeParser;
-  }
-  return newServiceContext;
-}
@@ -12,7 +12,6 @@ import {
  SentenceSplitter,
 } from "@llamaindex/core/node-parser";
 import { AsyncLocalStorage } from "@llamaindex/env";
-import type { ServiceContext } from "./ServiceContext.js";

 export type PromptConfig = {
  llm?: string;
@@ -163,42 +162,4 @@ class GlobalSettings implements Config {
  }
 }

-export const llmFromSettingsOrContext = (serviceContext?: ServiceContext) => {
-  if (serviceContext?.llm) {
-    return serviceContext.llm;
-  }
-
-  return Settings.llm;
-};
-
-export const nodeParserFromSettingsOrContext = (
-  serviceContext?: ServiceContext,
-) => {
-  if (serviceContext?.nodeParser) {
-    return serviceContext.nodeParser;
-  }
-
-  return Settings.nodeParser;
-};
-
-export const embedModelFromSettingsOrContext = (
-  serviceContext?: ServiceContext,
-) => {
-  if (serviceContext?.embedModel) {
-    return serviceContext.embedModel;
-  }
-
-  return Settings.embedModel;
-};
-
-export const promptHelperFromSettingsOrContext = (
-  serviceContext?: ServiceContext,
-) => {
-  if (serviceContext?.promptHelper) {
-    return serviceContext.promptHelper;
-  }
-
-  return Settings.promptHelper;
-};
-
 export const Settings = new GlobalSettings();
@@ -1,7 +0,0 @@
-import { AnthropicAgent } from "@llamaindex/anthropic";
-import { withContextAwareness } from "./contextAwareMixin.js";
-
-export const AnthropicContextAwareAgent = withContextAwareness(AnthropicAgent);
-export type { ContextAwareConfig } from "./contextAwareMixin.js";
-
-export * from "@llamaindex/anthropic";
@@ -1,7 +1,3 @@
-import {
-  AnthropicAgent,
-  type AnthropicAgentParams,
-} from "@llamaindex/anthropic";
 import type {
  NonStreamingChatEngineParams,
  StreamingChatEngineParams,
@@ -20,29 +16,21 @@ export interface ContextAwareState {
  retrievedContext: string | null;
 }

-export type SupportedAgent = typeof OpenAIAgent | typeof AnthropicAgent;
-export type AgentParams<T> = T extends typeof OpenAIAgent
-  ? OpenAIAgentParams
-  : T extends typeof AnthropicAgent
-    ? AnthropicAgentParams
-    : never;
+// TODO: support any LLMAgent
+export type SupportedAgent = typeof OpenAIAgent;
+export type AgentParams = OpenAIAgentParams;

 /**
 * ContextAwareAgentRunner enhances the base AgentRunner with the ability to retrieve and inject relevant context
 * for each query. This allows the agent to access and utilize appropriate information from a given index or retriever,
 * providing more informed and context-specific responses to user queries.
 */
-export function withContextAwareness<T extends SupportedAgent>(Base: T) {
+export function withContextAwareness(Base: SupportedAgent) {
  return class ContextAwareAgent extends Base {
    public readonly contextRetriever: BaseRetriever;
    public retrievedContext: string | null = null;
-    declare public chatHistory: T extends typeof OpenAIAgent
-      ? OpenAIAgent["chatHistory"]
-      : T extends typeof AnthropicAgent
-        ? AnthropicAgent["chatHistory"]
-        : never;

-    constructor(params: AgentParams<T> & ContextAwareConfig) {
+    constructor(params: AgentParams & ContextAwareConfig) {
      super(params);
      this.contextRetriever = params.contextRetriever;
    }
@@ -1,21 +1,6 @@
 export * from "@llamaindex/core/agent";
-export {
-  OllamaAgent,
-  OllamaAgentWorker,
-  type OllamaAgentParams,
-} from "@llamaindex/ollama";
-export {
-  AnthropicAgent,
-  AnthropicAgentWorker,
-  AnthropicContextAwareAgent,
-  type AnthropicAgentParams,
-} from "./anthropic.js";
-export {
-  OpenAIAgent,
-  OpenAIAgentWorker,
-  OpenAIContextAwareAgent,
-  type OpenAIAgentParams,
-} from "./openai.js";
+
+export { OpenAIContextAwareAgent } from "./openai.js";
 export {
  ReACTAgentWorker,
  ReActAgent,
@@ -1,5 +1,3 @@
-import type { ServiceContext } from "../ServiceContext.js";
-
 export type ClientParams = {
  apiKey?: string | undefined;
  baseUrl?: string | undefined;
@@ -9,5 +7,4 @@ export type CloudConstructorParams = {
  name: string;
  projectName: string;
  organizationId?: string | undefined;
-  serviceContext?: ServiceContext | undefined;
 } & ClientParams;
@@ -1 +0,0 @@
-export * from "@llamaindex/clip";
@@ -1 +0,0 @@
-export * from "@llamaindex/deepinfra";
@@ -1 +0,0 @@
-export { GEMINI_EMBEDDING_MODEL, GeminiEmbedding } from "@llamaindex/google";
@@ -1 +0,0 @@
-export * from "@llamaindex/huggingface";
@@ -1,4 +0,0 @@
-export {
-  MistralAIEmbedding,
-  MistralAIEmbeddingModelType,
-} from "@llamaindex/mistral";
@@ -1,4 +0,0 @@
-export {
-  MixedbreadAIEmbeddings,
-  type MixedbreadAIEmbeddingsParams,
-} from "@llamaindex/mixedbread";
@@ -1 +0,0 @@
-export { OllamaEmbedding } from "@llamaindex/ollama";
@@ -1,12 +1,5 @@
 export * from "@llamaindex/core/embeddings";
-export { ClipEmbedding, ClipEmbeddingModelType } from "./ClipEmbedding.js";
-export { DeepInfraEmbedding } from "./DeepInfraEmbedding.js";
 export { FireworksEmbedding } from "./fireworks.js";
-export { GEMINI_EMBEDDING_MODEL, GeminiEmbedding } from "./GeminiEmbedding.js";
-export * from "./HuggingFaceEmbedding.js";
 export * from "./JinaAIEmbedding.js";
-export * from "./MistralAIEmbedding.js";
-export * from "./MixedbreadAIEmbeddings.js";
-export { OllamaEmbedding } from "./OllamaEmbedding.js";
 export * from "./OpenAIEmbedding.js";
 export { TogetherEmbedding } from "./together.js";
@@ -18,8 +18,7 @@ import {
  messagesToHistory,
  streamReducer,
 } from "@llamaindex/core/utils";
-import type { ServiceContext } from "../../ServiceContext.js";
-import { llmFromSettingsOrContext } from "../../Settings.js";
+import { Settings } from "../../Settings.js";

 /**
 * CondenseQuestionChatEngine is used in conjunction with a Index (for example VectorStoreIndex).
@@ -44,7 +43,6 @@ export class CondenseQuestionChatEngine extends BaseChatEngine {
  constructor(init: {
    queryEngine: BaseQueryEngine;
    chatHistory: ChatMessage[];
-    serviceContext?: ServiceContext;
    condenseMessagePrompt?: CondenseQuestionPrompt;
  }) {
    super();
@@ -53,7 +51,7 @@ export class CondenseQuestionChatEngine extends BaseChatEngine {
    this.memory = new ChatMemoryBuffer({
      chatHistory: init?.chatHistory,
    });
-    this.llm = llmFromSettingsOrContext(init?.serviceContext);
+    this.llm = Settings.llm;
    this.condenseMessagePrompt =
      init?.condenseMessagePrompt ?? defaultCondenseQuestionPrompt;
  }
@@ -9,10 +9,9 @@ import {
 } from "@llamaindex/core/response-synthesizers";
 import { EngineResponse, type NodeWithScore } from "@llamaindex/core/schema";
 import { extractText } from "@llamaindex/core/utils";
-import type { ServiceContext } from "../../ServiceContext.js";
-import { llmFromSettingsOrContext } from "../../Settings.js";
 import type { BaseSelector } from "../../selectors/index.js";
 import { LLMSingleSelector } from "../../selectors/index.js";
+import { Settings } from "../../Settings.js";

 type RouterQueryEngineTool = {
  queryEngine: BaseQueryEngine;
@@ -60,7 +59,6 @@ export class RouterQueryEngine extends BaseQueryEngine {
  constructor(init: {
    selector: BaseSelector;
    queryEngineTools: RouterQueryEngineTool[];
-    serviceContext?: ServiceContext | undefined;
    summarizer?: BaseSynthesizer | undefined;
    verbose?: boolean | undefined;
  }) {
@@ -106,20 +104,16 @@ export class RouterQueryEngine extends BaseQueryEngine {
  static fromDefaults(init: {
    queryEngineTools: RouterQueryEngineTool[];
    selector?: BaseSelector;
-    serviceContext?: ServiceContext;
    summarizer?: BaseSynthesizer;
    verbose?: boolean;
  }) {
-    const serviceContext = init.serviceContext;
-
    return new RouterQueryEngine({
      selector:
        init.selector ??
        new LLMSingleSelector({
-          llm: llmFromSettingsOrContext(serviceContext),
+          llm: Settings.llm,
        }),
      queryEngineTools: init.queryEngineTools,
-      serviceContext,
      summarizer: init.summarizer,
      verbose: init.verbose,
    });
@@ -2,7 +2,6 @@ import type { BaseSynthesizer } from "@llamaindex/core/response-synthesizers";
 import { getResponseSynthesizer } from "@llamaindex/core/response-synthesizers";
 import { TextNode, type NodeWithScore } from "@llamaindex/core/schema";
 import { LLMQuestionGenerator } from "../../QuestionGenerator.js";
-import type { ServiceContext } from "../../ServiceContext.js";

 import type { BaseTool, ToolMetadata } from "@llamaindex/core/llms";
 import type { PromptsRecord } from "@llamaindex/core/prompts";
@@ -93,7 +92,6 @@ export class SubQuestionQueryEngine extends BaseQueryEngine {
    queryEngineTools: BaseTool[];
    questionGen?: BaseQuestionGenerator;
    responseSynthesizer?: BaseSynthesizer;
-    serviceContext?: ServiceContext;
  }) {
    const questionGen = init.questionGen ?? new LLMQuestionGenerator();
    const responseSynthesizer =
@@ -2,8 +2,7 @@ import type { ChatMessage, LLM } from "@llamaindex/core/llms";
 import { PromptMixin } from "@llamaindex/core/prompts";
 import { MetadataMode } from "@llamaindex/core/schema";
 import { extractText } from "@llamaindex/core/utils";
-import type { ServiceContext } from "../ServiceContext.js";
-import { llmFromSettingsOrContext } from "../Settings.js";
+import { Settings } from "../Settings.js";
 import type { CorrectnessSystemPrompt } from "./prompts.js";
 import {
  defaultCorrectnessSystemPrompt,
@@ -18,7 +17,6 @@ import type {
 import { defaultEvaluationParser } from "./utils.js";

 type CorrectnessParams = {
-  serviceContext?: ServiceContext;
  scoreThreshold?: number;
  parserFunction?: (str: string) => [number, string];
 };
@@ -35,7 +33,7 @@ export class CorrectnessEvaluator extends PromptMixin implements BaseEvaluator {
  constructor(params?: CorrectnessParams) {
    super();

-    this.llm = llmFromSettingsOrContext(params?.serviceContext);
+    this.llm = Settings.llm;
    this.correctnessPrompt = defaultCorrectnessSystemPrompt;
    this.scoreThreshold = params?.scoreThreshold ?? 4.0;
    this.parserFunction = params?.parserFunction ?? defaultEvaluationParser;
@@ -1,7 +1,6 @@
 import { PromptMixin, type ModuleRecord } from "@llamaindex/core/prompts";
 import { Document, MetadataMode } from "@llamaindex/core/schema";
 import { extractText } from "@llamaindex/core/utils";
-import type { ServiceContext } from "../ServiceContext.js";
 import { SummaryIndex } from "../indices/summary/index.js";
 import type {
  FaithfulnessRefinePrompt,
@@ -22,19 +21,16 @@ export class FaithfulnessEvaluator
  extends PromptMixin
  implements BaseEvaluator
 {
-  private serviceContext?: ServiceContext | undefined;
  private raiseError: boolean;
  private evalTemplate: FaithfulnessTextQAPrompt;
  private refineTemplate: FaithfulnessRefinePrompt;

  constructor(params?: {
-    serviceContext?: ServiceContext | undefined;
    raiseError?: boolean | undefined;
    faithfulnessSystemPrompt?: FaithfulnessTextQAPrompt | undefined;
    faithFulnessRefinePrompt?: FaithfulnessRefinePrompt | undefined;
  }) {
    super();
-    this.serviceContext = params?.serviceContext;
    this.raiseError = params?.raiseError ?? false;

    this.evalTemplate =
@@ -92,9 +88,7 @@ export class FaithfulnessEvaluator

    const docs = contexts?.map((context) => new Document({ text: context }));

-    const index = await SummaryIndex.fromDocuments(docs, {
-      serviceContext: this.serviceContext,
-    });
+    const index = await SummaryIndex.fromDocuments(docs, {});

    const queryEngine = index.asQueryEngine();

@@ -1,7 +1,6 @@
 import { PromptMixin, type ModuleRecord } from "@llamaindex/core/prompts";
 import { Document, MetadataMode } from "@llamaindex/core/schema";
 import { extractText } from "@llamaindex/core/utils";
-import type { ServiceContext } from "../ServiceContext.js";
 import { SummaryIndex } from "../indices/summary/index.js";
 import type { RelevancyEvalPrompt, RelevancyRefinePrompt } from "./prompts.js";
 import {
@@ -16,14 +15,12 @@ import type {
 } from "./types.js";

 type RelevancyParams = {
-  serviceContext?: ServiceContext | undefined;
  raiseError?: boolean | undefined;
  evalTemplate?: RelevancyEvalPrompt | undefined;
  refineTemplate?: RelevancyRefinePrompt | undefined;
 };

 export class RelevancyEvaluator extends PromptMixin implements BaseEvaluator {
-  private serviceContext?: ServiceContext | undefined;
  private raiseError: boolean;

  private evalTemplate: RelevancyEvalPrompt;
@@ -32,7 +29,6 @@ export class RelevancyEvaluator extends PromptMixin implements BaseEvaluator {
  constructor(params?: RelevancyParams) {
    super();

-    this.serviceContext = params?.serviceContext;
    this.raiseError = params?.raiseError ?? false;
    this.evalTemplate = params?.evalTemplate ?? defaultRelevancyEvalPrompt;
    this.refineTemplate =
@@ -78,9 +74,7 @@ export class RelevancyEvaluator extends PromptMixin implements BaseEvaluator {

    const docs = contexts?.map((context) => new Document({ text: context }));

-    const index = await SummaryIndex.fromDocuments(docs, {
-      serviceContext: this.serviceContext,
-    });
+    const index = await SummaryIndex.fromDocuments(docs, {});

    const queryResponse = `Question: ${extractText(query)}\nResponse: ${response}`;

@@ -32,7 +32,6 @@ export {
  DEFAULT_CONTEXT_WINDOW,
  DEFAULT_DOC_STORE_PERSIST_FILENAME,
  DEFAULT_GRAPH_STORE_PERSIST_FILENAME,
-  DEFAULT_IMAGE_VECTOR_NAMESPACE,
  DEFAULT_INDEX_STORE_PERSIST_FILENAME,
  DEFAULT_NAMESPACE,
  DEFAULT_NUM_OUTPUTS,
@@ -83,7 +82,6 @@ export * from "./OutputParser.js";
 export * from "./postprocessors/index.js";
 export * from "./QuestionGenerator.js";
 export * from "./selectors/index.js";
-export * from "./ServiceContext.js";
 export * from "./storage/StorageContext.js";
 export * from "./tools/index.js";
 export * from "./types.js";
@@ -1,21 +1,9 @@
 export * from "./index.edge.js";
-export * from "./readers/index.js";
-export * from "./storage/index.js";
-// Exports modules that doesn't support non-node.js runtime
-export {
-  HuggingFaceEmbedding,
-  HuggingFaceEmbeddingModelType,
-} from "./embeddings/HuggingFaceEmbedding.js";

-export {
-  GeminiVertexSession,
-  type VertexGeminiSessionOptions,
-} from "@llamaindex/google";
-
-// Expose AzureDynamicSessionTool for node.js runtime only
-export { AzureDynamicSessionTool } from "@llamaindex/azure";
+// TODO: clean up, move to jinaai package
 export { JinaAIEmbedding } from "./embeddings/JinaAIEmbedding.js";

-// Don't export vector store modules for non-node.js runtime on top level,
+// Don't export file-system stores for non-node.js runtime on top level,
 //  as we cannot guarantee that they will work in other environments
+export * from "./storage/index.js";
 export * from "./vector-store.js";
@@ -1,16 +1,18 @@
+import type {
+  BaseChatEngine,
+  ContextChatEngineOptions,
+} from "@llamaindex/core/chat-engine";
 import type { BaseQueryEngine } from "@llamaindex/core/query-engine";
 import type { BaseSynthesizer } from "@llamaindex/core/response-synthesizers";
 import type { BaseRetriever } from "@llamaindex/core/retriever";
 import type { BaseNode, Document } from "@llamaindex/core/schema";
 import type { BaseDocumentStore } from "@llamaindex/core/storage/doc-store";
 import type { BaseIndexStore } from "@llamaindex/core/storage/index-store";
-import type { ServiceContext } from "../ServiceContext.js";
-import { nodeParserFromSettingsOrContext } from "../Settings.js";
 import { runTransformations } from "../ingestion/IngestionPipeline.js";
+import { Settings } from "../Settings.js";
 import type { StorageContext } from "../storage/StorageContext.js";

 export interface BaseIndexInit<T> {
-  serviceContext?: ServiceContext | undefined;
  storageContext: StorageContext;
  docStore: BaseDocumentStore;
  indexStore?: BaseIndexStore | undefined;
@@ -22,14 +24,12 @@ export interface BaseIndexInit<T> {
 * they can be retrieved for our queries.
 */
 export abstract class BaseIndex<T> {
-  serviceContext?: ServiceContext | undefined;
  storageContext: StorageContext;
  docStore: BaseDocumentStore;
  indexStore?: BaseIndexStore | undefined;
  indexStruct: T;

  constructor(init: BaseIndexInit<T>) {
-    this.serviceContext = init.serviceContext;
    this.storageContext = init.storageContext;
    this.docStore = init.docStore;
    this.indexStore = init.indexStore;
@@ -53,15 +53,20 @@ export abstract class BaseIndex<T> {
    responseSynthesizer?: BaseSynthesizer;
  }): BaseQueryEngine;

+  /**
+   * Create a new chat engine from the index.
+   * @param options
+   */
+  abstract asChatEngine(
+    options?: Omit<ContextChatEngineOptions, "retriever">,
+  ): BaseChatEngine;
+
  /**
   * Insert a document into the index.
   * @param document
   */
  async insert(document: Document) {
-    const nodes = await runTransformations(
-      [document],
-      [nodeParserFromSettingsOrContext(this.serviceContext)],
-    );
+    const nodes = await runTransformations([document], [Settings.nodeParser]);
    await this.insertNodes(nodes);
    await this.docStore.setDocumentHash(document.id_, document.hash);
  }
@@ -5,8 +5,6 @@ import type {
  NodeWithScore,
 } from "@llamaindex/core/schema";
 import { MetadataMode } from "@llamaindex/core/schema";
-import type { ServiceContext } from "../../ServiceContext.js";
-import { serviceContextFromDefaults } from "../../ServiceContext.js";
 import { RetrieverQueryEngine } from "../../engines/query/index.js";
 import type { StorageContext } from "../../storage/StorageContext.js";
 import { storageContextFromDefaults } from "../../storage/StorageContext.js";
@@ -34,13 +32,17 @@ import type {
 import { BaseRetriever } from "@llamaindex/core/retriever";
 import type { BaseDocumentStore } from "@llamaindex/core/storage/doc-store";
 import { extractText } from "@llamaindex/core/utils";
-import { llmFromSettingsOrContext } from "../../Settings.js";
+import { Settings } from "../../Settings.js";
+import {
+  ContextChatEngine,
+  type BaseChatEngine,
+  type ContextChatEngineOptions,
+} from "../../engines/chat/index.js";

 export interface KeywordIndexOptions {
  nodes?: BaseNode[];
  indexStruct?: KeywordTable;
  indexId?: string;
-  serviceContext?: ServiceContext;
  llm?: LLM;
  storageContext?: StorageContext;
 }
@@ -79,7 +81,7 @@ abstract class BaseKeywordTableRetriever extends BaseRetriever {
    this.index = index;
    this.indexStruct = index.indexStruct;
    this.docstore = index.docStore;
-    this.llm = llmFromSettingsOrContext(index.serviceContext);
+    this.llm = Settings.llm;

    this.maxKeywordsPerQuery = maxKeywordsPerQuery;
    this.numChunksPerQuery = numChunksPerQuery;
@@ -152,6 +154,10 @@ const KeywordTableRetrieverMap = {
  [KeywordTableRetrieverMode.RAKE]: KeywordTableRAKERetriever,
 };

+export type KeywordTableIndexChatEngineOptions = {
+  retriever?: BaseRetriever;
+} & Omit<ContextChatEngineOptions, "retriever">;
+
 /**
 * The KeywordTableIndex, an index that extracts keywords from each Node and builds a mapping from each keyword to the corresponding Nodes of that keyword.
 */
@@ -163,7 +169,6 @@ export class KeywordTableIndex extends BaseIndex<KeywordTable> {
  static async init(options: KeywordIndexOptions): Promise<KeywordTableIndex> {
    const storageContext =
      options.storageContext ?? (await storageContextFromDefaults({}));
-    const serviceContext = options.serviceContext;
    const { docStore, indexStore } = storageContext;

    // Setup IndexStruct from storage
@@ -210,7 +215,6 @@ export class KeywordTableIndex extends BaseIndex<KeywordTable> {
      indexStruct = await KeywordTableIndex.buildIndexFromNodes(
        options.nodes,
        storageContext.docStore,
-        serviceContext,
      );

      await indexStore.addIndexStruct(indexStruct);
@@ -218,7 +222,6 @@ export class KeywordTableIndex extends BaseIndex<KeywordTable> {

    return new KeywordTableIndex({
      storageContext,
-      serviceContext,
      docStore,
      indexStore,
      indexStruct,
@@ -251,11 +254,16 @@ export class KeywordTableIndex extends BaseIndex<KeywordTable> {
    );
  }

-  static async extractKeywords(
-    text: string,
-    serviceContext?: ServiceContext,
-  ): Promise<Set<string>> {
-    const llm = llmFromSettingsOrContext(serviceContext);
+  asChatEngine(options?: KeywordTableIndexChatEngineOptions): BaseChatEngine {
+    const { retriever, ...contextChatEngineOptions } = options ?? {};
+    return new ContextChatEngine({
+      retriever: retriever ?? this.asRetriever(),
+      ...contextChatEngineOptions,
+    });
+  }
+
+  static async extractKeywords(text: string): Promise<Set<string>> {
+    const llm = Settings.llm;

    const response = await llm.complete({
      prompt: defaultKeywordExtractPrompt.format({
@@ -271,19 +279,16 @@ export class KeywordTableIndex extends BaseIndex<KeywordTable> {
   * @param documents
   * @param args
   * @param args.storageContext
-   * @param args.serviceContext
   * @returns
   */
  static async fromDocuments(
    documents: Document[],
    args: {
      storageContext?: StorageContext;
-      serviceContext?: ServiceContext;
    } = {},
  ): Promise<KeywordTableIndex> {
-    let { storageContext, serviceContext } = args;
+    let { storageContext } = args;
    storageContext = storageContext ?? (await storageContextFromDefaults({}));
-    serviceContext = serviceContext ?? serviceContextFromDefaults({});
    const docStore = storageContext.docStore;

    await docStore.addDocuments(documents, true);
@@ -291,11 +296,10 @@ export class KeywordTableIndex extends BaseIndex<KeywordTable> {
      await docStore.setDocumentHash(doc.id_, doc.hash);
    }

-    const nodes = serviceContext.nodeParser.getNodesFromDocuments(documents);
+    const nodes = Settings.nodeParser.getNodesFromDocuments(documents);
    const index = await KeywordTableIndex.init({
      nodes,
      storageContext,
-      serviceContext,
    });
    return index;
  }
@@ -304,20 +308,17 @@ export class KeywordTableIndex extends BaseIndex<KeywordTable> {
   * Get keywords for nodes and place them into the index.
   * @param nodes
   * @param docStore
-   * @param serviceContext
   * @returns
   */
  static async buildIndexFromNodes(
    nodes: BaseNode[],
    docStore: BaseDocumentStore,
-    serviceContext?: ServiceContext,
  ): Promise<KeywordTable> {
    const indexStruct = new KeywordTable();
    await docStore.addDocuments(nodes, true);
    for (const node of nodes) {
      const keywords = await KeywordTableIndex.extractKeywords(
        node.getContent(MetadataMode.LLM),
-        serviceContext,
      );
      indexStruct.addNode([...keywords], node.id_);
    }
@@ -328,7 +329,6 @@ export class KeywordTableIndex extends BaseIndex<KeywordTable> {
    for (const node of nodes) {
      const keywords = await KeywordTableIndex.extractKeywords(
        node.getContent(MetadataMode.LLM),
-        this.serviceContext,
      );
      this.indexStruct.addNode([...keywords], node.id_);
    }
@@ -19,11 +19,12 @@ import type {
 } from "@llamaindex/core/storage/doc-store";
 import { extractText } from "@llamaindex/core/utils";
 import _ from "lodash";
-import type { ServiceContext } from "../../ServiceContext.js";
-import {
-  llmFromSettingsOrContext,
-  nodeParserFromSettingsOrContext,
-} from "../../Settings.js";
+import { Settings } from "../../Settings.js";
+import type {
+  BaseChatEngine,
+  ContextChatEngineOptions,
+} from "../../engines/chat/index.js";
+import { ContextChatEngine } from "../../engines/chat/index.js";
 import { RetrieverQueryEngine } from "../../engines/query/index.js";
 import type { StorageContext } from "../../storage/StorageContext.js";
 import { storageContextFromDefaults } from "../../storage/StorageContext.js";
@@ -44,11 +45,15 @@ export enum SummaryRetrieverMode {
  LLM = "llm",
 }

+export type SummaryIndexChatEngineOptions = {
+  retriever?: BaseRetriever;
+  mode?: SummaryRetrieverMode;
+} & Omit<ContextChatEngineOptions, "retriever">;
+
 export interface SummaryIndexOptions {
  nodes?: BaseNode[] | undefined;
  indexStruct?: IndexList | undefined;
  indexId?: string | undefined;
-  serviceContext?: ServiceContext | undefined;
  storageContext?: StorageContext | undefined;
 }

@@ -63,7 +68,6 @@ export class SummaryIndex extends BaseIndex<IndexList> {
  static async init(options: SummaryIndexOptions): Promise<SummaryIndex> {
    const storageContext =
      options.storageContext ?? (await storageContextFromDefaults({}));
-    const serviceContext = options.serviceContext;
    const { docStore, indexStore } = storageContext;

    // Setup IndexStruct from storage
@@ -120,7 +124,6 @@ export class SummaryIndex extends BaseIndex<IndexList> {

    return new SummaryIndex({
      storageContext,
-      serviceContext,
      docStore,
      indexStore,
      indexStruct,
@@ -131,11 +134,9 @@ export class SummaryIndex extends BaseIndex<IndexList> {
    documents: Document[],
    args: {
      storageContext?: StorageContext | undefined;
-      serviceContext?: ServiceContext | undefined;
    } = {},
  ): Promise<SummaryIndex> {
    let { storageContext } = args;
-    const serviceContext = args.serviceContext;
    storageContext = storageContext ?? (await storageContextFromDefaults({}));
    const docStore = storageContext.docStore;

@@ -144,15 +145,11 @@ export class SummaryIndex extends BaseIndex<IndexList> {
      await docStore.setDocumentHash(doc.id_, doc.hash);
    }

-    const nodes =
-      nodeParserFromSettingsOrContext(serviceContext).getNodesFromDocuments(
-        documents,
-      );
+    const nodes = Settings.nodeParser.getNodesFromDocuments(documents);

    const index = await SummaryIndex.init({
      nodes,
      storageContext,
-      serviceContext,
    });
    return index;
  }
@@ -193,6 +190,16 @@ export class SummaryIndex extends BaseIndex<IndexList> {
    );
  }

+  asChatEngine(options?: SummaryIndexChatEngineOptions): BaseChatEngine {
+    const { retriever, mode, ...contextChatEngineOptions } = options ?? {};
+    return new ContextChatEngine({
+      retriever:
+        retriever ??
+        this.asRetriever({ mode: mode ?? SummaryRetrieverMode.DEFAULT }),
+      ...contextChatEngineOptions,
+    });
+  }
+
  static async buildIndexFromNodes(
    nodes: BaseNode[],
    docStore: BaseDocumentStore,
@@ -306,7 +313,6 @@ export class SummaryIndexLLMRetriever extends BaseRetriever {
  choiceBatchSize: number;
  formatNodeBatchFn: NodeFormatterFunction;
  parseChoiceSelectAnswerFn: ChoiceSelectParserFunction;
-  serviceContext?: ServiceContext | undefined;

  constructor(
    index: SummaryIndex,
@@ -314,7 +320,6 @@ export class SummaryIndexLLMRetriever extends BaseRetriever {
    choiceBatchSize: number = 10,
    formatNodeBatchFn?: NodeFormatterFunction,
    parseChoiceSelectAnswerFn?: ChoiceSelectParserFunction,
-    serviceContext?: ServiceContext,
  ) {
    super();
    this.index = index;
@@ -323,7 +328,6 @@ export class SummaryIndexLLMRetriever extends BaseRetriever {
    this.formatNodeBatchFn = formatNodeBatchFn || defaultFormatNodeBatchFn;
    this.parseChoiceSelectAnswerFn =
      parseChoiceSelectAnswerFn || defaultParseChoiceSelectAnswerFn;
-    this.serviceContext = serviceContext || index.serviceContext;
  }

  async _retrieve(query: QueryBundle): Promise<NodeWithScore[]> {
@@ -337,7 +341,7 @@ export class SummaryIndexLLMRetriever extends BaseRetriever {
      const fmtBatchStr = this.formatNodeBatchFn(nodesBatch);
      const input = { context: fmtBatchStr, query: extractText(query) };

-      const llm = llmFromSettingsOrContext(this.serviceContext);
+      const llm = Settings.llm;

      const rawResponse = (
        await llm.complete({
@@ -1,3 +1,7 @@
+import {
+  ContextChatEngine,
+  type ContextChatEngineOptions,
+} from "@llamaindex/core/chat-engine";
 import { IndexDict, IndexStructType } from "@llamaindex/core/data-structs";
 import {
  DEFAULT_SIMILARITY_TOP_K,
@@ -20,16 +24,15 @@ import {
 import type { BaseIndexStore } from "@llamaindex/core/storage/index-store";
 import { extractText } from "@llamaindex/core/utils";
 import { VectorStoreQueryMode } from "@llamaindex/core/vector-store";
-import type { ServiceContext } from "../../ServiceContext.js";
-import { nodeParserFromSettingsOrContext } from "../../Settings.js";
+import { Settings } from "../../Settings.js";
 import { RetrieverQueryEngine } from "../../engines/query/RetrieverQueryEngine.js";
 import {
  addNodesToVectorStores,
  runTransformations,
 } from "../../ingestion/IngestionPipeline.js";
 import {
-  DocStoreStrategy,
  createDocStoreStrategy,
+  DocStoreStrategy,
 } from "../../ingestion/strategies/index.js";
 import type { StorageContext } from "../../storage/StorageContext.js";
 import { storageContextFromDefaults } from "../../storage/StorageContext.js";
@@ -48,7 +51,6 @@ interface IndexStructOptions {
 }
 export interface VectorIndexOptions extends IndexStructOptions {
  nodes?: BaseNode[] | undefined;
-  serviceContext?: ServiceContext | undefined;
  storageContext?: StorageContext | undefined;
  vectorStores?: VectorStoreByType | undefined;
  logProgress?: boolean | undefined;
@@ -59,6 +61,12 @@ export interface VectorIndexConstructorProps extends BaseIndexInit<IndexDict> {
  vectorStores?: VectorStoreByType | undefined;
 }

+export type VectorIndexChatEngineOptions = {
+  retriever?: BaseRetriever;
+  similarityTopK?: number;
+  preFilters?: MetadataFilters;
+} & Omit<ContextChatEngineOptions, "retriever">;
+
 /**
 * The VectorStoreIndex, an index that stores the nodes only according to their vector embeddings.
 */
@@ -71,7 +79,7 @@ export class VectorStoreIndex extends BaseIndex<IndexDict> {
    super(init);
    this.indexStore = init.indexStore;
    this.vectorStores = init.vectorStores ?? init.storageContext.vectorStores;
-    this.embedModel = init.serviceContext?.embedModel;
+    this.embedModel = Settings.embedModel;
  }

  /**
@@ -84,7 +92,6 @@ export class VectorStoreIndex extends BaseIndex<IndexDict> {
  ): Promise<VectorStoreIndex> {
    const storageContext =
      options.storageContext ?? (await storageContextFromDefaults({}));
-    const serviceContext = options.serviceContext;
    const indexStore = storageContext.indexStore;
    const docStore = storageContext.docStore;

@@ -103,7 +110,6 @@ export class VectorStoreIndex extends BaseIndex<IndexDict> {

    const index = new this({
      storageContext,
-      serviceContext,
      docStore,
      indexStruct,
      indexStore,
@@ -204,10 +210,7 @@ export class VectorStoreIndex extends BaseIndex<IndexDict> {
    } = {},
  ): Promise<VectorStoreIndex> {
    args.storageContext =
-      args.storageContext ??
-      (await storageContextFromDefaults({
-        serviceContext: args.serviceContext,
-      }));
+      args.storageContext ?? (await storageContextFromDefaults({}));
    args.vectorStores = args.vectorStores ?? args.storageContext.vectorStores;
    args.docStoreStrategy =
      args.docStoreStrategy ??
@@ -230,7 +233,7 @@ export class VectorStoreIndex extends BaseIndex<IndexDict> {
    );
    args.nodes = await runTransformations(
      documents,
-      [nodeParserFromSettingsOrContext(args.serviceContext)],
+      [Settings.nodeParser],
      {},
      { docStoreStrategy },
    );
@@ -245,10 +248,7 @@ export class VectorStoreIndex extends BaseIndex<IndexDict> {
    }
  }

-  static async fromVectorStores(
-    vectorStores: VectorStoreByType,
-    serviceContext?: ServiceContext,
-  ) {
+  static async fromVectorStores(vectorStores: VectorStoreByType) {
    if (!vectorStores[ModalityType.TEXT]?.storesText) {
      throw new Error(
        "Cannot initialize from a vector store that does not store text",
@@ -262,20 +262,13 @@ export class VectorStoreIndex extends BaseIndex<IndexDict> {
    const index = await this.init({
      nodes: [],
      storageContext,
-      serviceContext,
    });

    return index;
  }

-  static async fromVectorStore(
-    vectorStore: BaseVectorStore,
-    serviceContext?: ServiceContext,
-  ) {
-    return this.fromVectorStores(
-      { [ModalityType.TEXT]: vectorStore },
-      serviceContext,
-    );
+  static async fromVectorStore(vectorStore: BaseVectorStore) {
+    return this.fromVectorStores({ [ModalityType.TEXT]: vectorStore });
  }

  asRetriever(
@@ -309,6 +302,25 @@ export class VectorStoreIndex extends BaseIndex<IndexDict> {
    );
  }

+  /**
+   * Convert the index to a chat engine.
+   * @param options The options for creating the chat engine
+   * @returns A ContextChatEngine that uses the index's retriever to get context for each query
+   */
+  asChatEngine(options: VectorIndexChatEngineOptions = {}) {
+    const {
+      retriever,
+      similarityTopK,
+      preFilters,
+      ...contextChatEngineOptions
+    } = options;
+    return new ContextChatEngine({
+      retriever:
+        retriever ?? this.asRetriever({ similarityTopK, filters: preFilters }),
+      ...contextChatEngineOptions,
+    });
+  }
+
  protected async insertNodesToStore(
    newIds: string[],
    nodes: BaseNode[],
@@ -407,7 +419,6 @@ export class VectorIndexRetriever extends BaseRetriever {
  index: VectorStoreIndex;
  topK: TopKMap;

-  serviceContext?: ServiceContext | undefined;
  filters?: MetadataFilters | undefined;
  queryMode?: VectorStoreQueryMode | undefined;

@@ -415,7 +426,6 @@ export class VectorIndexRetriever extends BaseRetriever {
    super();
    this.index = options.index;
    this.queryMode = options.mode ?? VectorStoreQueryMode.DEFAULT;
-    this.serviceContext = this.index.serviceContext;
    if ("topK" in options && options.topK) {
      this.topK = options.topK;
    } else {
@@ -1 +0,0 @@
-export * from "@llamaindex/anthropic";
@@ -1 +0,0 @@
-export * from "@llamaindex/deepinfra";
@@ -1 +0,0 @@
-export * from "@llamaindex/google";
@@ -1 +0,0 @@
-export * from "@llamaindex/groq";
@@ -1 +0,0 @@
-export * from "@llamaindex/huggingface";
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
Alex Yang	65958b2c2f	ci: fix	2025-02-12 21:12:41 +08:00
Alex Yang	6f3bf1b91f	chore: package.json repository url	2025-02-12 21:05:34 +08:00
Marcus Schiesser	f4588bc770	chore: Remove readers package from llamaindex (#1649 )	2025-02-12 17:16:41 +07:00
Marcus Schiesser	b49037612d	remove service context (#1618 ) Co-authored-by: thucpn <thucsh2@gmail.com>	2025-02-12 15:10:11 +07:00
Thuc Pham	a87efb91a4	docs: update chat engine docs (#1648 )	2025-02-12 12:45:26 +07:00
Thuc Pham	6a4a73760b	chore: remove re-exporting packages in llamaindex (#1624 ) Co-authored-by: Marcus Schiesser <mail@marcusschiesser.de>	2025-02-12 12:44:52 +07:00
Alex Yang	1564831158	ci: fix pkg-pr-new release (#1646 )	2025-02-11 22:42:42 +08:00
Alex Yang	4d94f6e50d	test: smoke test with cjs/esm dual package (#1644 )	2025-02-11 15:02:06 +08:00
Thuc Pham	7bd5d9340c	docs: update workflow doc (#1637 ) Co-authored-by: Marcus Schiesser <mail@marcusschiesser.de>	2025-02-11 13:41:11 +07:00
Thuc Pham	d924c63162	feat: asChatEngine function for index (#1640 )	2025-02-11 12:57:15 +07:00
				`@@ -1 +0,0 @@`
				`export { GEMINI_EMBEDDING_MODEL, GeminiEmbedding } from "@llamaindex/google";`
				`@@ -1 +0,0 @@`
				`export { OllamaEmbedding } from "@llamaindex/ollama";`